-

Fishery Studies of Ribb River, Lake !Ib

@b Tana Basin, Ethiopia #!b

* m El 573 b

!!B Final Report VOL. 2 lib-

Presented to the World Bank-financed Ethiopian-Nile

Irrigation and Drainage Project Coordination Office,

Ministry of Water Resources

BY:

Abebe Getahun (PhD) ------LeadA quatic Ecologist

Eshete Dejen (PhD) ------- Aquatic Ecologist

Wassie Anteneh (MSc) ------- Aquatic Ecologist

January 2008

Addis Ababa, Ethiopia

Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized

Ribb River Fishery Studies: Final Report January 2008

Table of Contents

Content

Table of contents

List of Tables

List of Figures

Page

i

11

. . .

111

Acknowledgments iv

Acronyms vi

Executive Summary 1

Background 5

Ethiopian Drainage basins 5

Ethiopian Freshwater fishes 5

Lake Tana 7

Lake Tana fishes and fisheries 8

Migration in fishes 9

The effects of dam building on fishes 10

The initiation and execution of this study 13

Objectives of the study 14

Study sites, Materials and Methods 14

Study sites 14

Materials and Methods 19

Results and Discussions 22

Fish species diversity from Ribb River and its tributaries 22

Fish species abundance in the Ribb River and its tributaries 2 5

Length-weight relationship of the dominant Labeobarbus spp. 2 8

Role of Labeobarbus spp in the food web of Lake Tana 3 0

Gonad Maturity status and migration behavior of Labeobarbus spp. 3 1

Spawning Segregations 3 6

Spatial segregation 36

Temporal segregation 38

Suitable spawning habitats in the Ribb River and its tributaries 3 8

Farmer's use of fish from Ribb River and its tributaries 43

The contribution of Ribb fishes to Lake Tana fish ecology and production 44

Species composition of Labeobarbus spp. from Lake Tana 46

Ribb River Fishery Studies: Final Report January 2008

Fish catch trends from Lake Tana 4 8

Fish production and marketing from Lake Tana 5 0

The nature and number of Lake Tana fishers, their bases and organizations 5 1

Fishery Policy and Management 54

Major Findings, Conclusions and Recommended Mitigation Measures 56

Major Findings and Conclusions 5 6

Recommended Mitigation Measures 59

References 62

Appendices 67

Appendix I: Names of contacted individuals and Institutions 67

Appendix 11: Questionnaire for the socio-economic studies 68

Appendix 111: Pictures showing activities 69

Appendix IV: Raw data collected 7 8

Ribb River Fishery Studies: Final Report January 2008

List of Tables

Table Content Page

Table 1. Average value of oxygen, temperature, pH and depth at the sampling sites. 18

Table 2. Gonad maturity stages and descriptions for cyprinids 2 1

Table 3. The species of fish identified from the different sampling sites of Ribb Ever 24

Table 4. Occurrence and total number of species of fish from the different sampling 24

sites of Ribb River.

Table 5. Total abundance of species from all the sampling sites of Ebb River 26

Table 6. Number of species and their percentage composition fiom the different

sampling sites of Ribb River 27

Table 7. Spatial distribution of gonad stages of the Labeobarbus spp. of Lake Tana 34

spawning in Ribb River

Table 8. Tributaries of Ribb River and the distance fiom their origin to the main Channel 40

Table 9. Species number and percentage composition pooled from all monthly 47

samples per site from January 2000 to December 2004 (BAFLRC, 2007)

Table 10. Species composition of the Labeobarbus species in Lake Tana 48

Table 11. Fish catch data in tons for Motorized and Reed Boat fishery in Lake Tana 49

Table 12. Fish prices at Bahir Dar 5 1

Table 13. Dependents on the fishery sector that are directly involved in fishing and post 54

harvest processing

Ribb River Fishery Studies: Final Report January 2008

List of Figures

Figure Content Page

Fig. 1. Ribb River water shed and the sampling sites 15

Fig. 2. Length-weight relationship of Labeobarbus species of Lake Tana

spawning in Ribb River 29

Fig. 3. Simplified food web of Lake Tana

Fig. 4. Maturity stages in the most common Labeobarbus spp.

Fig. 5. Temporal variation of gonad stages of Labeobarbus of Lake Tana

in Ribb River 3 5

Fig. 6. Spatial segregation of Labeobarbus species of Lake Tana spawning

in Ribb River.

Fig. 7. The planned dam site on Ribb River and the reservoir area

Ribb River Fishery Studies: Final Report January 2008

Acknowledgments :

We thank the Ministry of Water Resources (World Bank Financed Ethiopian-Nile

Imgation & Drainage Project Coordination Office) for the great cooperation we

received during our entire working period. We are thankful to the Bahirdar Fisheries

and Other Living Aquatic Resources Research Center, the Amhara Regional

Agricultural Research Institute, Ministry of Water Resources (Amhara Region), South

Gondar Zone Administration, Libo Kemkem Woreda Agriculture Office, Ebenat

Woreda Agriculture Office and several other institutions for providing us most needed

information and logistics.

Fishery experts (Belay Abdissa, Goraw Goshu and Mesfin Tsegaw) and a number of

fishermen (Gebresellassie Meseret, Getnet Temesgen, Ayenew Gediff, Tiruneh Belay,

Asrate Wondie, Endalamaw Asres) were involved in the sampling effort and we are

indebted to all of them. We are thankful to the able drivers, Asnake Aynekulu, Habte

Muche and h a r e that helped us in very difficult conditions. We are also thankful to

several farmers and local residents that have assisted the work in providing

information and animals and transporting our equipments. We also thank Geremew

Mellese, Abebe Ameha and Fitsum Abebe for their contributions in preparing the

Ribb watershed map, organizing data, formatting and other editorial assistance.

Ribb River Fishery Studies: Final Report January 2008

Acronyms:

ACU:

BA:

BAFLRC:

BRE:

CH:

CPUE:

CRA:

DEG:

ETB:

FL:

FPME:

GAR:

GON:

HA:

INT :

KI

KE:

LON:

LTFRDP:

MAC:

MC1:

MC2:

MC3 :

MC4:

MC5:

ME :

MEG:

NED:

PLA:

RM:

SH:

SUR:

TIL:

Labeobarbus acutrostris

Barya River (tributary of Ribb River)

Bahir Dar Fish and other Living Resources Research Center

Labeobarbus brevicephalus

Chibirna River (tributary of Ribb River)

Catch Per Unit Effort

Labeobarbus crassibarbis

Labeobarbus degeni

Ethiopian Birr

Fork Length

Fish Production and Marketing Enterprise

Garra sp.

Labeobarbus gorgorensis

Hamus River (tributary of Ribb River)

Labeobarbus intermedius

Kirarign River (tributary of Ribb River)

Keha River (tributary of Ribb River)

Labeobarbus longissirnus

Lake Tana Fisheries Resource Development Program

Labeobarbus macrolepidotus

Main Ribb River at the Bahir Dar-Gondar Bridge near Woreta town

Main Ribb River at the old Broken Bridge (on the old road from Zeha

to Debretabor)

Main Ribb River at the dam site

Main Ribb River above the junction with Hamus River

Main Ribb River above the junction with Melo River

Melo River (tributary of Ribb River)

Labeobarbus megastoina

Labeobarbus nedgia

Labeobarbus platydorsus

River mouth

Shini River (tributary of Ribb River)

Labeobarbus surkis

Oreochromis niloticus

vi

Ribb River Fishery Studies: Final Report January 2008

TRU: Labeobarbus truttiformis

TSA: Labeobarbus tsanensis

TW: Total Weight

VAR: Varicorhinus beso

vii

Ribb River Fishery Studies: Final Report January 2008

EXECUTIVE SUMMARY

A dam is planned to be constructed on the Ribb River some 50 km north east of Lake

Tana. This study was designed to investigate the dam's effect on fishes of

Labeobarbus spp. that migrate from Lake Tana to Ribb River and its tributaries for

spawning. 13 sampling sites were identified at the river mouth, on the main Ribb

River and its tributaries below and above the dam site. Fishes were collected from

each sampling site in three separate field trips during the main rainy season (end of

July to beginning of November). A total of 2457 fish specimens (1287 females, 1143

males and 27 unidentified specimens) were collected using gill nets of different mesh

sizes (6, 8, 10, and 12 cm stretched mesh size), hooks and lines, fykes, cast nets, traps

and scoop nets and fishes were also purchased from the local fishers for biological

investigation.

Eighteen species belonging to five genera and three families were represented in the

catch. L. intermedius, L. brevicephalus, L. megastoma, L. truttiformis and L. tsanensis

were the most abundant species migrating to Ribb River and its tributaries. These

Labeobarbus species were found to aggregate at the Ribb River mouth before the

onset of migration and they are found to migrate to all the tributaries and up to the

most distant site sampled (Main channel above the junction with Melo River located

some 15 km above the dam).

Most of the Labeobarbus specimens collected during the study period were having

gonads that were mature and running; from the total of 2075 Labeobarbus specimens,

887 specimens were mature, 835 specimens were running and 53 were spent. There is

no evidence for specificity of habitat at which each species of Labeobarbus spawn

and hence spatial segregation is out-ruled. However, there is evidence that they

segregate temporally except in the case of L. internzedius. Among the dominant

species, L. megastoma was the first to aggregate at the river mouth and run to

upstream followed by L. truttiformis and L. tsanensis. The last to aggregate and

migrate upstream is L. brevicephalus.

All the tributaries and the upstream Ribb main channel can be considered as suitable

breeding habitats for Labeobarbus spp. The total estimated length of spawning habitat

along the main Ribb River and its tributaries is 351.5 km. The area below the dam

contributes 57.3 % while the spawning habitat above the dam contributes 42.7 % of

1

Ribb River Fishery Studies: Final Report January 2008

the total. Moreover it has to be noted that higher relative abundance of Labeobarbus

spp. was observed from sampling sites below the dam than above the dam.

It has been estimated that 48 % of the fish production from Lake Tana that landed at

the Southern Gulf (about 1000 tons) is contributed by the fish caught from the Ribb

River mouth and its surroundings. This is about 480 tons contributed by mainly Nile

tilapia, cat fish and Labeobarbus spp. The contribution of Labeobarbus spp. is 20 %

of the total catch, which is about 100 tons per year. It is reasonable to estimate, thus,

that about 60 tons (57.3%) could be from fishes that spawn below the dam and 40

tons (42.7%) fish production could come from fishes that spawn above the dam.

Therefore, the inability for the fishes to migrate past the dam site could bring about an

estimated loss of about 40 tons of fish production per year from Lake Tana.

Moreover, if proper management measures are not taken on the main river and the

tributaries below the dam site, and the water level of the main river and its tributaries

fall below the level required for the migration of the species as well as inundating the

wetlands, it could result in an estimated loss of about 480 tons of fish per year from

Lake Tana. This is because any negative measures on the wetlands and the main Ribb

River could affect the migratory species as well as the production of Nile tilapia and

catfish that have their main breeding grounds on the wetlands. In monetary terms 480

tons is about Birr 1440000lyear (producer's rate) or Birr 5280000lyear (retailer's

rate).

The above estimate is based on the amount of catch that landed at the southern gulf

which is one-tenth of the estimated potential of Lake Tana (estimated potential is

about 10000 tons). Therefore, the potential monetary loss that could be accrued to

Lake Tana fish production due to damages on the Ribb River and its tributaries could

roughly be estimated to Birr 50280000 per year.

The number of fish that are migrating to and from Ribb River and its tributaries could

be estimated from the actual production of Lake Tana that is contributed by

Labeobarbus spp. from Ribb River mouth and sourrounding areas which is about 100

tons (100,000 kg) per year. It can reasonably be suggested, thus, that about 200000

adult individuals (an individual adult weighing, on the average, about 500 grams) of

Labeobarbus species migrate to and from Ribb River. Of this, about 115000

Ribb River Fishery Studies: Final Report January 2008

individuals may spawn below the dam while about 75000 individuals spawn above

the dam.

The river mouth and shallow areas around Ribb River are ideal breeding and feeding

grounds for Nile Tilapia (Oreochromis rziloticus) and African cat fish (Clarias

gariepinus) and hence are important habitats that play great roles in the functioning of

Lake Tana ecosystem.

Ribb River fishery is insignificant contributing less than 1% to the total fish catch in

the Lake Tana basin; a total of about 360 quintals (or 36 tons) of fish is estimated to

be harvested from the river annually. Therefore, there are no communities whose

livelihoods entirely depend on the fishery of the river.

The contribution of Labeobarbus species to Lake Tana fishery was about 40 % in the

1990s and declined to 17 % in 2000s. This implies that this species flock is under

pressure from different sources of threat that include fishing at the river mouth during

breeding season (recruitment overfishing), spawning habitat destruction, silt load and

similar other factors.

With the formation of the reservoir a new fishery is expected to emerge. The reservoir

is estimated to be about 1918 ha. On average, the fish production from such type of

tropical reservoir is estimated to be about 100 kghalyr. Therefore, 1918 quintal of

fish could be produced annually from this reservoir.

Generally, the dam will have an impact on the migratory Labeobarbus species that

spawn past the dam in upstream Ribb River and its tributaries and that will, in turn,

have some impact on the fishery of Lake Tana. It could also have effect on the

quantity of water that will be available for the migratory fishes as well as for

inundating the wetlands. However, the impact will be insignificant and tolerable if the

following major mitigation measures are taken:

> The time of the dam closure1 blocking should not coincide with the period of

migration of the fishes (the main rainy season).

> A sufficient volume of flow is required to inundate flood plains, recharge

wetlands and provide sufficient depth of water for larger species. Further

3

Ribb River Fishery Studies: Final Report January 2008

hydrological studies may be needed to determine the amount of water required

to inundate flood plains and recharge wetlands. However, about 20 cm height

of water is the minimum requirement to keep larger species migrating to and

from Ribb River.

> It is important to protect the watershed of the main river channel and the

tributaries from further deforestation and environmental degradation so that

the diversity and productivity of the fish will be maintained. It is

recommended that the riparian zone of the main Ribb River and the tributaries

is afforested.

> Water use for irrigation from the small tributaries should be minimized

especially during the breeding season. Other habitat destruction activities

should be avoided.

> In order to improve the livelihood of the people around the dam area, fish

must be stocked into the reservoir and the fish species that will be stocked

should be species from Lake Tana propagated at a hatchery. Introduction of

any new fish species into the reservoir will affect the fish diversity of the Lake

Tana sub-basin.

> The fish stock from Lake Tana is sharply declining even before the

construction of the dam. It is, therefore, important to rehabilitate the stock

using artificial propagation.

> Farmers fishing from Ribb River use destructive fishing methods (poisoning

and blocking), and this need to be strongly banned.

o Fishing should be closed during the spawning months (July to October).

o Fishers must use gillnets of 10 cm and above stretched mesh size.

o Licensing of fishers must be immediately materialized.

o Enforcement of management measures, effective training and extension

work should incorporate active participation of the fisher community

including the upstream and downstream communities.

Finally, it is hoped that the dam will serve the intended purpose of irrigating farm

lands and ultimately mitigating the poverty level and misery of the surrounding

farmers without severely compromising the natural ecosystem, which is related to

the wellbeing of several generations to come.

Ribb River Fishery Studies: Final Report January 2008

BACKGROUND

Ethiopian Drainage Basins

Much of Africa is arid, and Ethiopia could perhaps be called the "water tower of

eastern Africa". The country is endowed with some 7000 km2 of standing water and

some 7000 km length of flowing water. Based on similarities of the fauna (especially

the fish fauna) and following the model of freshwater ecoregions of Africa (Thieme et

al., 2006) the freshwater systems of Ethiopia can be conveniently placed under 5

freshwater ecoregions. These are:

The Ethiopian Highlands (includes streams, rivers and lakes in the highlands

of Ethiopia, but excluding Lake Tana).

Lake Tana (because of its unique fish fauna).

Northern Rift (rift valley lakes excluding Lakes Abaya and Chamo because of

the Nilo-Sudanic affinities of their fish fauna)

Lake Turkana (includes the Omo River and its tributaries as well as Lakes

Abaya and Chamo)

Shebele Juba catchments (includes tributaries of Wabi Shebele, Genale, Dawa,

and Fafan).

Red Sea coastal (the Awash system and the saline lakes of northern Ethiopia

that includes Lakes Abbe, Afambo, Afdera, and Asale)

These freshwater ecoregions can further be divided into drainage basins. The drainage

pattern in Ethiopia is the result of the uplifting during the Tertiary period, which

created the Rift Valley and consequently the two separate highlands (Mohr, 1966;

Westphal, 1975). Since water bodies found in one drainage basin are somehow

interconnected, similarity in their biota is evident. According to Mesfin Woldemariam

in Shibru Tedla (1973), the Ethiopian freshwater system can be classified into seven

drainage basins. These are the Abay, Awash, Baro Akobo, Omo-Gibe, Rift Lakes,

Tekeze and Wabi Shebele-Genale basins.

Ethiopian Freshwater Fishes

The freshwater fish fauna of Ethiopia is of particular interest since it contains a

mixture of Nilo-Sudanic, East African, and endemic forms (Roberts, 1975; Abebe

Getahun and Stiassny, 1998). The Nilo-Sudanic forms are represented by a large

number of species found in the Baro-Akobo, Omo-Gibe, and Abay drainage basins

Ribb River Fishery Studies: Final Report January 2008

(e.g. members of the genera Alestes, Bagrus, Citharinus, Hydrocynus, Hyperopisus,

Labeo, Mormyrus etc.). The southern Rift valley (Lakes Abaya and Chamo), and the

Shebele-Genale basins also have elements of these forms. It is believed that these

lakes and river basins had former connections with the upper White Nile (through

Lake Rudolf in the former case) as recently as 7500 years ago (Roberts, 1975). These

Nilo-Sudanic forms are related to West African fishes and this too is believed to be

due to past connections of the Nile to Central and West African river systems

(Boulenger, 1905: Nichols and Griscom, 19 17; Nichols, 1928).

The highland east African forms are found in the northern Rift Valley lakes (e.g.

Lakes Awassa, Ziwai, Langano), the highland lakes (e.g. Tana and Hayq), and

associated river systems, and the Awash drainage basin. These include members of

the genera Barbus, Labeobarbus, Clarias, Garra, Oreochromis, and Varicorhinus.

They are related to fishes of eastern, northern and southern Africa. Some elements are

shared with waters of western Africa. For example, G. dembeensis is a widely

distributed cyprinid species found in 6 countries (Ethiopia, Kenya, Egypt, Tanzania,

Cameroun and Nigeria). Nilotic fishes are almost entirely absent from the Awash and

northern rift valley lakes.

Although extensive review work is currently in progress, it appears that a preliminary

listing of about 152 valid indigenous species represents what is so far known from

Ethiopian freshwaters. There are additionally 10 exotic species. Of the 152 indigenous

species, about 39 species and two sub-species are endemic to Ethiopia. Moreover, the

inadequacies of the present study underline the contention that further extensive

collections and identifications will raise both the total number and the number of

endemic species of the country.

The highest species diversity is recorded from Baro basin, followed by Abay, Rifi

Lakes, Wabi Shebele and Omo-Gibe basins. It appears that this high diversity is partly

attributable to the presence of highly diverse and rich habitats, but probably also to

relatively high level of exploration and collections done in these relatively accessible

water bodies. However, endemicity seems to be highest in Abay and Awash basins.

This is due to the endemic "species flock" of Lake Tana and the presence of some

endemic fishes adapted to localized habitats in small streams in the highlands of north

and central Ethiopia. Lake Tana has 28 species and one sub species of which 20

6

Ribb River Fishery Studies: Final Report January 2008

species and one sub species are Ethiopian endemics. 18 species are endemic to Lake

Tana.

The drainage basins that are rich in species like the Baro and Omo-Gibe contribute an

insignificant proportion of the country's endemic fauna. Only one endemic species

(Nemacheilus abyssinicus) has so far been recorded from these drainages and this

species has also been recorded from Lake Tana. Low levels of endemism are probably

due to the Baro and Omo-Gibe drainage basins having connections (present and past)

with the Nile and west and central African river systems and as a result all the fish

fauna represent widespread Nilo-Sudanic forms.

The major commercially important fish species of the country include Oreochronzis

niloticus, Labeobarbus spp., Lutes niloticus, Clarias gariepinus, Bagrus docmak, and

Cyprinus carpio (introduced).

Lake Tana

Lake Tana is the largest lake in Ethiopia, with a surface area of 3200 ~ r annd a~

watershed of 16500 ~ r loncat~ed a t 1830 m above sea level. It forms the headwaters

of the Blue Nile, which carries more than 80% of the total volume of the Nile River at

Khartoum, Sudan. The lake has been isolated from the lower Blue Nile basin by a 40

m, high water fall, 30 km downstream from the Blue Nile outflow. Lake Tana

emerged as one of the global top 250 lake regions most important for biological

diversity (Barker, 2004).

Lake Tana is an oligo-mesotrophic shallow lake with an average depth of 8 m and

maximum depth of 14 m (Wassie Anteneh, 2005). The lake is turbid, well mixed and

has no thermocline (Eshete Dejen et al, 2004). Fogera (on the east) and Dembea (on

the north) plains border major parts of Lake Tana, and they are considered to be the

buffering zones of the lake (Nagelkerke, 1997). The lake is believed to have

originated two million years ago by volcanic blocking of the Blue Nile River (Mohr,

1962). It assumed its present shape through blocking of a 50 km long quaternary

basalt flow, which filled the exit channel of the Blue Nile River (Chorowicz et al.,

1998). However, there are strong evidences that Lake Tana had dried up between

16000 and 50000 years ago (Lamp et al., 2004).

Ribb River Fishery Studies: Final Report January 2008

Lake Tana Fishes and Fisheries

In Lake Tana, the families Cichlidae and Clariidae are represented by only one

species each, Oreochromis niloticus and Clarias gariepinus, respectively.

Nemacheilus abyssinicus is an endemic species belonging to the family Balitoridae

and inhabit the littoral areas of Lake Tana. The largest fish family in the lake is

Cyprinidae, represented by four genera, Barbus, Garra, Varicorhinus and

Labeobarbus. The genus Barbus includes the "small" barbs and is represented by

three species, namely, B. humilis, B. pleurograma and B. tanapelagius (de Graff et al,

2000). Varicorhinus is represented by a single species, V. beso. The genus Garra is

represented by four species, G. dembecha, G. dembeensis, G. regressus and G. tana

(Stiassny and Abebe Getahun, 2007).

The most significant genus of the family Cyprinidae in Lake Tana is Labeobarbus.

The Labeobarbus species of Lake Tana have previously been classified under the

genus Barbus. However, large, hexaploid African Barbus are renamed as

Labeobarbus (Skelton, 2001, Berrebi and Tsigenpoulos, 2003, Snoeks, 2004). The

new genus name better reflects their phylogenetic distance from other members of the

overly lumped genus Barbus. Labeobarbus spp. differ not only in their resource

partitioning (feeding) but also in their reproductive strategies (de Graff et al., 2005).

There are 15 species of Labeobarbus forming a unique species flock in Lake Tana,

the only cyprind species flock in the world, after the ones in Lake Lanao vanished

because of overexploitation.

Until the end of the 1980s, fishing on Lake Tana was subsistence reed boat fishery. In

1986 motorized boats and nylon gill nets were introduced as part of the Lake Tana

Fisheries Resource Development Program, which was initiated by the Ethiopian

Ministry of Agriculture, the Ethiopian Orthodox Church, and two Dutch NGOs (ISEURK

and ICCO-Zeist) (Wassie Anteneh, 2005). Accordingly, the total annual catches

increased from 39 MT in 1987 to 360 MT in 1997 (Tesfaye Wudneh, 1998).

However, Catch per Unit Effort of the Labeobarbus species from the commercial gill

net fishery drastically dropped down from 63 kgltrip in 1991 to 28 kgltrip in 2001 (de

Graff et al., 2004). The same author has reported from the southern gulf of Lake Tana

about 75% decline (in biomass) and 80% (in number) of the Labeobarbus spp. (L.

acutrostris, L. macrophthalmus, L. platydorsus, L. brevicephalus, L. tsanensis, L.

intermedius). The most plausible explanation for the decline of the stock is not

Ribb River Fishery Studies: Final Report January 2008

natural environmental destruction but recruitment over fishing by the commercial gill

net fishery (de Graff et al., 2004) and poisoning of the spawning stock in rivers using

the crushed seeds of birbira (Milletia ferruginea) (Nagelkerke and Sibbing, 1996;

Abebe Ameha, 2004).

The commercial gill net fishery on Labeobarbus spp. is highly seasonal and mainly

targets the spawning aggregations, as more than 50% of the annual catch is obtained

in the river mouths during August and September.

Migration in Fishes

According to Rodriguez-Ruiz and Grando-Lorencio (1992) in Wassie Anteneh, 2005,

migration of fish refers to a displacement between two or more habitats, commonly

between feeding and reproduction habitats, with a regular periodicity (sometimes

annually) and involving a large fraction of a population. Control of the timing of

migration to rivers depends on interaction between the internal physiological state of

the fish and the external triggering factors in the environment (Northcote et al., 1970).

The external triggering factors include mostly moon phase, photoperiod, river flow,

water temperature, turbidity, and water volume.

Adults of anadromous species migrate up rivers to spawn and the young descend to

the lakes to feed and grow. Results of limited number of studies indicate that

migratory behavior of individual fish indicate that migratory movements are not

random, but are oriented, with varying precision, in the general direction of home

(Leggett, 1977). Several species of fish are known to be capable of obtaining

directional information from the sun, polarized light, and geomagnetic fields. Some

limited inertial guidance may also be involved.

An impressive body of literature supports the hypothesis that fish migrations involve a

continuous optimization of physiological and neurological states in response to a

multiplicity of environmental stimuli. Recognition of the home area apparently

involves both olfactory and local topographic cues. Each river, and apparently each

tributary, has a characteristic odor. Homing may result in reproductive isolation. This

isolation is essential to the development of complex behavioral, energetic, and

reproductive adaptations to the reproductive habitat occupied.

Ribb River Fishery Studies: Final Report January 2008

Gonad maturation is endogenously regulated but environmentally synchronized, in the

temperate region, most probably by the rate of photoperiod change (Thorpe, 1988).

However, most tropical freshwater fishes spawn seasonally during the rainy period

(Lowe-McConnell, 1975; Payne, 1986). Most large cyprinids of Africa spawn by

making a single annual breeding migration to upstream areas of rivers (Lowe

McConnell, 1975; Tomasson et al., 1984).

The Effects of Dam Building on Fishes

Dams block the migratory movements of spawning fishes in streams. The blockage of

fish movements upstream can have a very significant and negative impact on fish

biodiversity. According to McAllister et al.(unspecified date) many stocks of

Salmonidae and Clupeidae have been lost as a consequence. In the Columbia River,

U.S.A., more than 200 stocks of anadromous, Pacific salmonids became extinct as a

result of such actions.

While dramatic declines in migratory species such as lampreys, sturgeons, salmons

and clupeids were well known in European rivers, other fishes, the so called resident

or non-migratory fishes which perform in-stream movements require attention. These

include different species of minnows, sculpins, and graylings. Even small sized

species such as the white bream, Abranzis bjoerkna, were found to migrate up to 60

km from the place they were tagged.

Reservoirs fonned as a result of damming trap suspended particles, reducing turbidity

downstream. Many species are adapted to natural turbidity; for example, turbid water

catfishes have small eyes, refined senses of smell and touch in their sensitive barbells.

The turbid water helps conceal the fish and other biota from visual predators like

birds. When normally turbid water becomes clear below dams, the indigenous species

may find themselves at a disadvantage. Other animal species may move in, filter

feeders and aquatic vegetation may flourish. Sediment burrowing species may find

their habitat has diminished. Flood plain ecosystems and deltas may no longer be

replenished by the annual transport of sediment. Silt and increased turbidity, above

natural levels, can interfere with primary production.

Water quality, flow and seasonality of flow are not normally disrupted in the upstream

area above the reservoir so impacts are generally less than for the reservoir and

10

Ribb River Fishery Studies: Final Report January 2008

downstream areas. Nevertheless, the dam and the reservoir affect migratory

movements of species into and out of this upstream area. The genetic exchanges with

downstream segments is reduced or prevented.

In the construction of reservoirs, the clearing of vegetation, movement of earth and

rock, the presence of humans and machinery, bringing in construction materials, use

of explosives, noise and reducing or cutting off river flow and increasing turbidity,

will affect biodiversity. Removal of forests or other vegetation over a wide area,

excavation, earth and rock movement and reductions in river flow are the most

significant.

During reservoir filling the river and any associated wetland areas become inundated.

Riffles, runs and pools of the river are lost beneath the rising waters, leading to the

extirpation (or extinction) of habitat sensitive riverine species with tightly defined

niche requirements. Fishes in rivers are generally well adapted to flowing water. The

transformation of a river to a reservoir, therefore, poses a problem for the resident;

mainly riverine species that are not adapted to the new conditions.

Reservoir fisheries are one of the frequently claimed benefits of impoundments. The

changes in catches following impoundments are variable. However, the catches in

new reservoirs frequently go through a "boom and bust" cycle, with catches initially

increasing following filling of the reservoir and then declining. Therefore, impact

assessments of dams should be based on the long term catches.

In the downstream segment, most of the impacts of a dam are negative. In a

preliminary assessment of 66 case studies of the impact of dam construction on fishes,

based on qualitative information, 73% of the impacts were negative and only 27%

were positive. About 55% of the impacts were below the dam and linked to fish

migrations and to flood plain access (McAllister, et al. unspecified date).

Upstream impacts are generally less than those in the reservoir or downstream. The

exception to this generalization is the migratory species that move up and downstream

and use such movements to maintain genetic diversity.

It is clear that the World Charter for Nature was adopted by the UN General

Assembly in 1982. It provides the guiding principles that should govern human

11

Ribb River Fishery Studies: Final Report January 2008

responsibility for biodiversity. It states that "activities which might have an impact on

nature shall be controlled, and the best available technologies that minimize

significant risks to nature or adverse effects shall be used; in particular.

> Activities which are likely to cause irreversible damage to nature should be

avoided;

P Activities which are likely to pose a significant risk to nature shall be preceded

by an exhaustive examination; their proponents shall demonstrate that

expected benefits outweigh potential damage to nature, and where potential

adverse effects are not fully understood, the activities should not proceed;

> Activities which may disturb nature shall be preceded by assessment of their

consequences, and environmental impact studies of development projects shall

be constructed in advance, and if they are to be undertaken, such activities

shall be planned and camed out so as to minimize potential adverse impacts".

Similar statements are found in Convention on Biological Diversity. The Convention

on the Conservation of Migratory Species of Wild Animals (CMS), also states, among

its fundamental principles, that: 'the parties acknowledge the need to take actions to

avoid any migratory species becoming endangered" which is particularly relevant to

our report.

The World Bank's operational Policy on natural habitats requires that comprehensive

analysis should demonstrate that overall benefits from a project outweigh the

environmental costs before significant conversion of natural habitats is allowed,

unless there are no feasible alternatives for the project.

Although approximately 300 new freshwater species are discovered each year,

amphibians, fish and wetland birds are at high risk of becoming extinct in many

regions of the world. More than 20% of all freshwater fish species are now considered

threatened or endangered, mostly due to damming (Truffer, et al., 2003).

The Initiation and Execution of this Study

This study was initiated and sponsored by the Ministry of Water Resources to find out

whether or not the envisioned dam building on the main Ribb River, Lake Tana

tributary, will affect the migratory fishes of the Labeobarbus species and the extent of

its effect on the ecosystem and the people depending on the fishes for their livelihood.

Ribb River Fishery Studies: Final Report January 2008

Accordingly, a team was organized that consists of the lead aquatic ecologist and two

other aquatic ecologists together with fishery experts and fishermen. The first field

study and sampling took place from 31 July to 13 August 2007. In this first trip,

sampling sites were properly identified and fish samples, although few, were collected

from the main Ribb River and its tributaries.

As a continuation of this effort the second field trip to the main river and its tributaries

took place in two periods from 27 August to 3 September 2007 and from 17 to 30

September 2007. The Second field trip was divided into two periods because of the

heavy rains that continued until mid- September that made it very difficult to work on

the main Ribb River. In the first part (27 August to 3 September) of the trip only the

tributaries were sampled while in the second part both the main channel (Ribb) and its

tributaries were explored.

The third and final field trip took place between 20 October and 5 November 2007.

During this time the water level of the tributaries was so low that it was not possible

to sample from the tributaries, and hence sampling was mainly limited to the main

river channel and the river mouth. It has to be noted that the study was conducted in a

very difficult logistic and physical conditions and demanded mobilization of human

and material resources within a short period of time.

This final report is a compilation and analysis of data collected in those three field

trips from Ribb River and its tributaries. The report has an "Introduction" that

consists of general and brief background information on Ethiopian drainage basjns,

Ethiopian freshwater fishes, Lake Tana and its fish and fisheries, migration and dam

building. The "Materials and Methods" section describes the sampling sites as well

as the materials and methods used in the sampling process. The "Results and

Discussions" section presents the major findings and elaborates the implications of

these findings. The "Conclusions and Recommendations" outline the selected

outcomes of the report and indicate possible mitigation measures that need to be taken

before, during and after the construction of the dam. References and appendices

(consisting of individuals and institutions contacted, questionnaire prepared for socioeconomic

data collection, raw data and some selected pictures) are also part of the

report.

Ribb River Fishery Studies: Final Report January 2008

Objectives of the Study

The specific objectives of the study are to find out:

P whether or not there are migratory fish species from Lake Tana to Ribb River.

P the diversity and abundance of fish species that migrate to Ribb River from

Lake Tana.

P whether or not dam building on Ribb River will affect the migratory behavior

of fishes of Lake Tana.

P the extent of damage and habitat loss, otherwise important for the spawning

fishes, that would follow as a result of the dam construction,

P the extent of economic damage that may ensue on Lake Tana fisheries,

P mitigation measures that should be taken in order to minimize the negative

effects.

STUDY SITES, MATERIALS AND METHODS

Study Sites

Ribb River is approximately 90 km long and originates from Gunna Mountains range

in South Gondar Administrative Zone at an altitude of 2400 m. 13 sampling sites were

identified on the main Ribb River and its tributaries (Fig. 1). The sampling sites are

distributed evenly along much of the length of the Main River below and above the

dam site. A brief description of each site is given below and their physical conditions

are listed in Table 1.

Ribb River Fishery Studies: Final Report January 2008

Ribb River Fishery Studies: Final Report January 2008

Description of the Study Sites:

1. River mouth (RM): N 12'02'27.6" and E 37'35'49.3"; Elevation 1800 m.

This is a site located at the junction of Ribb River with Lake Tana. This is a

site where migratory fishes are expected to aggregate before the onset of

migration to the main river and its tributaries. This site is accessible only by

boat from the Bahir Dar Gulf.

2. Chibirna River (tributary of Ribb River) (CH): N 12O04'30.8" and E

37'44'4.7''; Elevation 1836 m. This is a temporary tributary of Ribb River

located between Addis Zemen and Yifag on the western side of the road

between Bahir Dar and Gondar. There is over flooding during the rainy season

and it is one of the major contributors to the flood over Fogera plains.

3. Main Ribb River at the Bahir Dar-Gondar Bridge near Woreta town

(MC1): N 11'59'38'' and E 37'42'38.6''; Elevation 1799 m. This is located at

a plain farm area with some eucalyptus trees around the banks. The

surrounding is highly exposed to flooding and erosion and the water at this site

is usually turbid.

4. Shini River (tributary of Ribb River) (SH): N 12~05'42.7" and E

037'45'36.5''; Elevation 1869 m. The above two tributaries, Shini and

Chibirna, originate from different widely separated localities and they join at

the Fogera plains of the Libo Kem Kem side and highly contribute to the

surrounding flood and finally join the lower part of Ribb River in the rainy

season. At the beginning of the study, we didn't consider Shini River as a

separate sampling site, since we thought that the two rivers join on the Fogera

plains, and may not show distinctions. However, we later found out that Shini

River, by its own, is a very good breeding habitat as its bed is filled with

gravels all along its upper length. So, we collected additional samples from the

river and included it in the list of sampling sites.

5. Kirarign River (tributary of Ribb River) (KI): N 12O02'04.8" and E

37047'32.2''; Elevation 1801 m. It measures to 2-5 m wide, with pool up to

1.50 m depth and the river mouth (the place where the tributary joins the main

Ribb channel) is muddy. Further upstream the river bed is composed of cobble

and gravel. The vegetation is quite scarce and there are farm lands all around.

6. Keha River (tributary of Ribb River) (KE): N 12'02'07.4'' and E

037O56'45.4"; Elevation 1836 m. It is about 3-5 m wide, with pools up to 2 m

Ribb River Fishery Studies: Final Report January 2008

deep. The river mouth is muddy whereas further upstream the river bed is

composed of gravels.

7. Main Ribb River at the old Broken Bridge (on the old road from Zeha to

Debretabor) (MC2): N 12'02'54.2'' and E 37'59'05.2".

This channel, during the rainy season, is very wide measuring to about 30

meters with depths ranging from 1.2 to 5 meters. It is an area denuded of its

natural vegetation and all the riparian area is tilled for growing crops. Hence,

the river is too turbid during the rainy seasons due to the introduction of silt

from the surrounding land through erosion.

8. Barya River (tributary of Ribb River) (BA): N 12'02'54.4'' and E

37'59'49.0''. This is a small tributary river of the main Ribb and flows,

although in small quantity, through out the year. There is farm land on all its

sides and the riparian vegetation has disappeared except at some gorgy areas

where tilling the land is difficult and some herbs are eminent.

9. Main Ribb River at the dam site (MC3): N 12'02'01.4'' and E

38'00'22.0''.

During the rainy season Ribb at the dam site is rapid. The bed is covered by

coble and pebbles. There are two hills (Aydagn on the northern Ebenat side

and Tigab Amba on the southern Farta side) in the middle of which Ribb

flows. The land around this site is used as a farm land for crop production.

There is little riparian vegetation left.

10. Hamus River (tributary of Ribb River) (HA): N12'00'26.1" and E

38"01'16.6". Hamus River is a perennial tributary which joins Ribb River

from the northern Ebenat Woreda side. Hamus River is the largest tributary of

Ribb River. During the sampling months, it is about 6-8 m wide and 1.5-2

meters deep. The velocity of the water at the mouth is higher. In August, it

was as turbid as the main Ribb channel, but in September and October the

water was very clear. The river bed has pebbles and gravels at its mouth.

Hamus River at its upper part passes through mountainous and highly

dissected land and at its mouth it is clear with no vegetation cover.

11. Main Ribb River above the junction with Hamus River (MC4):

N12"00'35" and E 38'02'34''. Ribb River at this site forms some pockets of

pools at some intervals of the stretch. The bottom profile of the river is

covered by boulders and pebbles. The velocity of the river is higher except at

the pockets. The land in the area is farmed for crop production. There is no

17

Ribb River Fishery Studies: Final Report January 2008

riparian vegetation except some eucalyptus trees. The water is turbid like that

of the lower sites of the Ribb River.

12. Melo River (tributary of Ribb River) (ME): N 12'02'32" and E

38'02'45.5''.

Melo River is a small temporary tributary stream which joins Ribb River on

the southern Farta Woreda side. It is about 3 m wide and 0.5 m deep during

the sampling period. It was less turbid as compared to the main Ribb River.

The bottom at its mouth is gravel bedded. Like that of Hamus River it is not

shaded with riparian vegetation or macrophytes. The water of Melo River is

slow flowing.

13. Main Ribb River above the junction with Melo River (MCS):

N12'02'15.3" and E 38'02'24.4''. Ribb River at this site flows relatively

slowly as the land is plain and it forms deeper pools. The water was clear as

compared to the lower sites. The bottom is covered by pebbles. There is no

vegetation cover. Reptiles such as the Nile crocodile were common at this site.

Table 1. Average value of oxygen, temperature, pH and depth at the sampling sites.

Depth (m)

3.5

0.5

2

0.3

0.75

0.72

1.25

0.5

1.25

0.75

0.74

1.24

1.23

pH

7.89

NA

7.9 1

8.24

7.61

8.09

7.79

7.86

7.5

7.76

7.78

7.80

7.83

Temp.

(OC )

22.1

23.1

22.1

22.7

18.3

23.1

21.2

22.6

23.8

22.3

22.9

21.5

22.3

Oxygen

(mgl-')

6.4

6.7

5.6

NA

NA

NA

6.5

6.6

6.8

6.9

7.12

6,2

6,3

Site

number

1

2

3

4

5

6

7

8

9

10

11

12

13

Site name

Ribb River Mouth

Chibirna River

Ribb main channel (Bahir Dar-

Gondar bridge)

Keha River

Kirarign

Shini

Ribb main channel (Ziha

Debretabor old bridge)

Barya River Mouth

Dam site

Hamus River

Mello River

Ribb main channel (Bet. Hamus

and Mello Rivers)

Ribb main channel (upper Mello

River)

Ribb River Fishery Studies: Final Report January 2008

MATERIALS AND METHODS

The samplings were extensively and systematically done throughout the rainy season

(end of July to beginning of November), collecting 2457 specimens of Labeobarbus

spp. and other groups of fish from 13 sampling sites.

9 Site selection was accomplished by inspection of the main river for

appropriate fishing sites and determining whether or not the tributaries flow

throughout the year. The information was secured through interview of experts

of the Ministry of Agriculture and the local fishermen and farmers.

9 In all of the sites day time and overnight gill net settings were made using

polyfilament gill nets with 6, 8, 10 and 12 cm stretched mesh size and with a

panel length of 25 and 50 meters and depth of 3 meters each. Fykes, cast nets,

scoop nets and hooks and lines were also employed. Fish were also purchased

in some upstream sampling sites from local fishermen who used locally made

scoop net and cast net to capture fish.

9 Fish collected in the river mouths were transported fresh to the laboratory of

Bahir Dar Fish and Other Aquatic Life Research Center whereas catches from

upstream sites were processed at the site.

9 All of the fishes caught were identified to species level with immediate

inspection (for obviously known species) and with the help of identification

key (Nagelkerke et al., 1994).

P Measurements of Fork length (to the nearest 0.1 cm), Total weight (to the

nearest 0.1 gram), and Gonad weight (to the nearest 0.01 gram) were taken

using measuring board and sensitive balances.

9 Each fish was dissected; the gonads were examined visually and sexed. The

gonad maturity stage of each Labeobarbus specimen was determined

according to Pet et al. (1996), modified from De Silva et al. (1985) in

Nagelkerke, 1997 (Table 2).

k Length-weight relationship of five most abundant Labeobarbus species of

Lake Tana spawning in Ribb River was computed using least square

regression analysis of TW = ~ F (BLag~ena l and Tesch, 1978), where, TW =

total weight (g), FL = fork length, a is the intercept, and b is the slope of the

regression line.

Ribb River Fishery Studies: Final Report January 2008

P Physico-chemical parameters were measured using GPS, Oxygen meter, pH

meter, and Conductivity meter. Depth was measured with Sechi Disc and

measuring rope. The type of bottom substratum and the status of the

surrounding vegetation were inspected and recorded.

P Farmers residing around the dam site and elsewhere around the sampling sites

were interviewed based on questions designed for this purpose (see appendix 2

for the type of questions).

P The Amhara Regional Bureau of the Ministry of Agriculture and Rural

development and the Fish and Other Living Aquatic Resources Research

Center at Bahir Dar were contacted and consulted for availability and

acquisition of long term fisheries data (both commercial and experimental).

P Libraries at the Science Faculty, Addis Ababa University and the Amhara

Regional Agricultural Research Institute were searched for relevant literature.

Ribb River Fishery Studies: Final Report January 2008

Table 2. Gonad maturity stages and descriptions for cyprinids (Source: Nagelkerke,

1997)

Gonad Male Female

stages

I Immature, impossible to distinguish Immature, impossible to distinguish

females from males. Gonads are a females from males. Gonads are a

pair of transparent strings running pair of transparent strings running

along the body cavity. along the body cavity.

Unambiguously male, very small Unambiguously female, very small

testes, white-reddish, not lobed, tube- ovaries, tube shaped and reddish, eggs

shaped strings not visible.

I11 Larger testes, white-reddish, some Ovary somewhat larger and starting

what lobed starting to flatten sideways to flatten sideways, eggs visible, but

very small

IV Large testes, white-reddish, lobed, Larger ovary, flattened sideways and

flattened sideways almost covering body cavity wall, eggs

yellowish

V Large, white testes, some sperm runs Larger and full ovary, completely

out when testis is cut covering body cavity wall, yellowish

eggs run out when ovary is cut

VI Large white testes, running, large Running, yellow eggs can be extruded

amount of sperm runs out when testis by puttingpressure on the abdomen

is cut

VII Spent, empty testes, reddish and Spent, wrinkled ovary, reddish,

wrinkled containing a few yellow eggs

Ribb River Fishery Studies: Final Report January 2008

RESULTS AND DISCUSSION

Fish Species Diversity from Ribb River and its Tributaries

Eighteen species belonging to five genera and three families were identified from the

thirteen sampling sites (Table 3). Of the total of 15 species of Labeobarbus occurring

in Lake Tana, 13 species (86.7 %) have been identified from the river mouth, main

Ribb River and its tributaries. The highest species diversity (1 6 species) has been

recorded from the river mouth where only Labeobarbus degeni and Garra sp. were

absent. The next two sites that showed higher diversity were the Ribb channel at the

Bridge on the main road from Bahir Dar to Gondar and Chibirna River. It is

interesting to note that these sites are the closest to the River mouth. It is reasonable,

thus, to see higher diversity at sites which are close to the origin of "dispersal",

although the diversity may be dependent on the intensity of sampling, the type of

gears used in sampling and the period of sampling.

Clarias gariepinus, 0. niloticus, V. beso, B. degni, and Garra species were caught

together with the migratory Labeobarbus species. Clarias gariepinus and 0. niloticus

were common at the Ribb River mouth and in the main channel but rare or absent in

the upper small tributaries of this river. Both are commonly found in most African

lakes, rivers and reservoirs as they are ecologically most resilient fishes. Both species

spawn in the floodplains and littoral parts of the lake (Zenebe Tadesse, 1997; Tesfaye

Wudneh, 1998) but never migrate to upstream rivers unlike Labeobarbus species.

Peak spawning for C. gariepinus occurs at the beginning of July while although 0.

niloticus spawns throughout the year, peak spawning occurs in March and July.

Therefore, the specimens of these two species collected in the Ribb upstream are most

probably dwelling in the river since the season and breeding ground of these species is

different from our sampling time and sites. This is again substantiated by the presence

of a large number of immature fish in the samples.

The remaining three cyprinids: V. beso, L. degni and Garra sp. were also collected

during the sampling period (Table 3). Varichorhinus beso, which contributes 1% of

the commercial catch in Lake Tana, was almost as abundant as some Labeobarbus

species such as L. tsanensis collected in this study. Only two specimens of V. beso

were caught at the Ribb River mouth. Most of the specimens of this species were

Ribb River Fishery Studies: Final Report January 2008

collected from the main channel and the small tributaries of Ribb River. Out of the

101 specimens 41 were immature. This species spawns throughout the year (Wassie

Anteneh, unpublished data). No published data is available about the reproductive

biology of this species, however, from gonad maturity status analysis and rare

occurrence of this species at the river mouth; it is possible to conclude that V. beso

lives (feeds and reproduces) in Ribb River basin.

The unverified species L. degeni was collected from Barya River and at the dam site.

No report or published data is available on the occurrence of this species in the Lake

Tana basin. Similarly unidentified (at the species level) specimens from the genus

Garra were collected in the upper tributary streams of Ribb River. This genus is

common in the Lake Tana basin (Wassie Anteneh, 2005; Akewak Geremew, 2007).

Generally, the diversity of Labeobarbus spp. observed in Ribb River and tributaries is

high. The construction of the dam is feard to create bottlenecks in some population of

the Laboebarbus spp. (those that are already in small numbers) and threaten them to

extinction.

Ribb River Fishery Studies: Final Report January 2008

Table 3. Species of fish identified fiom the different sampling sites of Ribb River.

Family Genus Species

Cyprinidae Labeobarbus acutirostris

Cyprinidae Labeobarbus brevicephalus

Cyprinidae Labeobarbus crassibarbis

fC rinidae i

Cyprinidae Labeobarbus gorgorensis

C yprinidae Labeobarbus intermedius

Cyprinidae Labeobarbus longissimus

Cyprinidae Labeobarbus macropthalmus

C yprinidae Labeobarbus megastoma

C yprinidae Labeobarbus nedgia

Cyprinidae Labeobarbus platydorsus

Cwrinidae Labeobarbus surkis

Cvprinidae Labeobarbus truttiformis

Cwrinidae Labeobarbus tsanensis

Cvprinidae Garra dembeensis

Cmrinidae Varicorhinus besso

Claridae

Cichlidae Oreochromis niloticus

Table 4. Occurrence and total number of species of fish from the different sampling

sites of Ribb River.

Sitel RM MCl CH SH KI KE MC2 BA MC3 HA MC4 ME MC5

Soecies

L. acutirostris J J

brevicephalus J J J J J J J J J J

crassibarbis J

gorgorensis J J J J d d d

intermedius J J J J J J J J J J J

longissirnus .\I

macropthalmus J

megastoma J J J J J J J J

nedgia J J J J J J J J J

platydorsus J J

surkis J d

truttiformis J J J J J J

tsanensis J J J J J

G. dembeensis .\/ J .\/ J

V. besso J J J J J J J J J J

C. gariepinus J J J J J J J J J J J

O. niloticus J J J J J

Total 16 11 10 7 7 8 9 8 6 8 4 4 6

Ribb River Fishery Studies: Final Report January 2008

Fish Species Abundance in the Ribb River and its Tributaries

The most abundant species from all sampling sites was Labeobarbus intermedius

constituting more than 30 % of the total number of specimens collected. This is not

surprising as this is a "waste basket", as all specimens that cannot easily be identified

to any distinct species are included in this species "complex". This species is also

believed to be the original (ancestral) group that is well adapted to riverine conditions.

L. brevicephalus, L. megastoma and L. truttiformis are other Labeobarbus spp. that

were found in relative abundance of 25.07 %, 9.28 % and 6.35 %, respectively.

Specimens of L. nedgia were relatively more abundant (about 25 specimensltrip) in

the upstream areas and L. nedgia with immature gonads were also caught in the

upstream sampling sites. Moreover, this species did not aggregate at the Ribb River

mouth and the most probable explanation is that L. nedgia may be feeding and

spawning in Ribb River basin and does not enter into the lake. This riverine dwelling

behavior of this species was also reported from Megech River basin (Wassie Anteneh,

2005).

In previous studies (Nagelkerke and Sibbing, 1996; Dgebuaze et al., 1999; Palstra et

al., 2004; de Graaf et al., 2005), and this one, seven species (L. crassibarbis, L.

dainellii, L. gorgorensis, L. gorguari, L. longissimus, L. nedgia, and L. surkis) did not

form aggregation in Gelgel Abbay, Gelda, Gumara, and Ribb River mouths.

Moreover, the rare occurrence of L. platydorsus, L. acutirostris and L.

macrophtalmus, in Ribb River in this study may be explained by the limited number

of samplings conducted. Experimental data taken from the different fishing sites of

Lake Tana (Table 10) indicate a similar pattern of abundance in the lake.

In general, two hypotheses can be forwarded for all the rare and missing Labeobarbus

species:

1. They may spawn in a smaller perennial river Arno-Garno (Fig. I), or

2. These fish species may spawn in the lake (lacustrine spawners).

From the other genera, 0. niloticus, C. gariepinus and V. beso were found in 5.98 %,

5.62 % and 4.19 %, respectively.

Ribb River Fishery Studies: Final Report January 2008

Table 5. Total abundance of species from all the sampling sites of Ribb River

Species Number of specimens % ~ o m ~ o s i t i o n

L. acutirostris 9 0.37

L. brevicephalus 616 25.07

L. crassibarbis 2 0.08

L. degeni 5 0.20

L. gorgorensis 78 3.17

L. macropthalmus 1 0.04

L. nedgia 59 2.40

L. platydorsus 57 2.32

L. surkis 10 0.42

L. tsanensis 62 2.52

G. dembeensis 10 0.4 1

V. besso 103 4.19

C. gariepinus 138 5.62

0. niloticus 147 5.98

Total 2457 100

The highest number of specimens (50.14% of the total) was collected from the River

Mouth site. This is a site where all migrating species aggregate before the onset of

migration. The main channel at the Bridge between Bahir Dar and Gondar town

stands second in abundance of species, whereas the tributary rivers, Chibirna, Barya,

Keha and Kirarign Rivers stood third, fourth, fifth and sixth, respectively. The

abundance of specimens, apparently, positively correlates with the diversity of species

except the case at MC2 (the main channel at the old Ziha-Debretabor Bridge) where

the diversity was relatively higher, but the abundance was relatively lower. In

principle, however, the diversity of species may not necessarily positively correlate

with the abundance of specimens. The abundance of specimens is apparently observed

at sites below the dam site and it is also evident that the extent of spawning habitat

(total length of Ribb and tributary rivers) is greater below the dam than above the

dam. Therefore, these habitas need proper management and monitoring.

Ribb River Fishery Studies: Final Report January 2008

Table 6. Number of species and their percentage composition from the different

sampling sites of Ribb River

Site/ RM MC 1 CH SH KI KE MC2

Species

N % N % N % N % N % N % N %

* acutirostris 6 .24 3 .12 -- -- -- -- -- -- -- -- --

brevicephal 133 5.4 -- -- 95 3.9 57 2.3 102 4.2 104 4.23 24 .98

crassibarbis 2 .08 -- -- -- -- -- -- -- -- -- -- -- --

degeni -- -- -- -- -- -- -- -- -- -- -- -- -- --

gorgorensis 27 1.1 36 1.5 3 .I2 -- -- -- -- 1 .04 2 .08

intermedius 405 16.5 168 6.8 16 .65 4 .16 3 .12 8 .33 22 .90

longissimus 1 .04 -- -- -- -- -- -- -- -- -- -- -- --

macropthal 1 .04 -- -- -- -- -- -- -- -- -- -- -- --

megastoma 144 5.9 10 .41 2 .08 -- -- 5 .20 16 .65 1 .04

nedgia 3 .12 1 .04 1 .12 2 .08 1 .04 6 .24 7 .28

platydorsus 55 2.24 2 .08 -- -- -- -- -- -- -- -- -- --

surkis 9 .37 1 .04 -- -- -- -- -- -- -- -- -- --

truttiformis 127 5.2 6 .24 9 .37 -- -- 7 .28 -- -- 2 .08

tsanensis 52 2.1 2 .08 -- -- -- -- -- -- 3 .I2 2 .08

dembeensis -- -- -- -- 5 .20 1 .04 -- -- -- -- -- --

besso 2 .08 -- -- 20 .81 7 .28 -- -- 2 .08 3 .12

gariepinus 4 1 1.7 48 2.0 3 .12 3 .12 5 .20 7 .28 5 .20

niloticus 124 5.1 4 .16 8 .33 8 .33 1 .04 -- -- -- --

Total 1132 50.14 281 11.4 162 6.6 82 3..34 124 5.05 147 5.98 68 2.77

Table 6-contd. Number of species and their percentage composition from the different

sampling sites of Ribb River.

Site/ Species B A MC3 HA MC4 ME MC5 Overall Total

N % N % N % N % N % N % N Yo

, acutirostris -- -- -- -- -- -- -- -- -- -- -- -- 9 .37

brevicephal 34 1.38 10 .41 24 .98 13 .53 9 .37 11 .45 616 25.1

crassibarbis -- -- -- -- -- -- -- -- -- -- -- -- 2 .08

degeni -- -- 5 .20 -- -- -- -- -- -- -- -- 5 .20

gorgorensis 5 .20 -- -- 4 .16 -- -- -- -- -- -- 78 3.18

intermedius 61 2.48 32 1.3 27 1.1 11 .45 3 .12 15 .61 775 31.54

longissimus -- -- -- -- -- -- -- -- -- -- -- -- 1 .04

macropthal -- -- -- -- -- -- -- -- -- -- -- -- 1 .04

megastoma 35 1.43 -- -- 15 .61 -- -- -- -- -- -- 228 9.28

nedgia 9 .37 14 .57 11 .45 -- -- 1 .04 3 .12 59 2.40

platydorsus -- -- -- -- -- -- -- -- -- -- -- -- 57 2.32

surkis -- -- -- -- -- -- -- -- -- -- -- -- 10 .4 1

tmttiformis -- -- -- -- 5 .20 -- -- -- -- -- -- 156 6.34

tsanensis 3 .12 -- -- -- -- -- -- -- -- -- -- 62 2.52

dembeensis -- -- -- -- 3 .12 -- -- 1 .04 -- -- 10 .41

besso 7 -- 40 1.63 15 .61 4 .16 -- -- 3 .12 103 4.19

gariepinus 4 18 -- -- 1 .04 -- -- 3 .12 138 5.61

niloticus -- -- -- -- -- -- -- -- -- -- 2 .08 147 5.97

Total 158 6.43 119 4.84 104 4.23 29 1.18 14 .57 37 1..51 2457 100

Ribb River Fishery Studies: Final Report January 2008

Length-weight Relationship of the Dominant Labeobarbus spp.

Total weight was curnilinearly related with fork length in the five most dominant

Labeobarbus species of Lake Tana spawning in Ribb River. The regression

coefficients were near the cube value (b=3). The results obtained in this study fit with

the "theoretical" cube law; which means growth in these fish species is isometric

(weight increases at a rate of about a cube of increase in length) (Fig. 2). A similar

result was obtained for these species by Wassie Anteneh (2005) in the samples taken

from Dirma and Megech tributary rivers of Lake Tana and Naglekerke et al. (1 994) in

the lacustrine samples of Lake Tana.

L brevicephalus 4

N= 646

444

*** 4

100 L 8

J 9 :

z= 0

ti 10 15 20 25 30 35 Bj w 3 1800 -

CC

TW = 0 . 0 0 9 1''~ ~ ~

4

1500-

L. intermedius

N=765

1200 -

900 -

600 -

300 -

0 - I I

Ribb River Fishery Studies: Final Report January 2008

Foi.1L~ al,$l~( FL) (all)

Fig. 2. Length-weight relationship of Labeobarbus species of Lake Tana spawning in

Ribb River.

Ribb River Fishery Studies: Final Report January 2008

Role of Labeobarbus spp in the food web of Lake Tana

We will specifically deal with the role and position of the Labeobarbus spp. in the

food web of Lake Tana because they are the ones that migrate and have direct relation

with the proposed dam.

Eight species of the fifteen endemic Labeobarbus spp. (more than 65 % of all

labeobarbs) are piscivorous: two are obligate piscivorous (L. acutirostris, L.

truttiforntis) and six facultative piscivorous (L. dainellii, L. gorguari, L. longissimus,

L. nzacrophtalmus, L. megastonta, L. platydorsus) (Sibbing & Nagelkerke, 2001).

Experiments showed that these piscivores are very clumsy predators, most probably

because they have a narrow pharyngeal slit and lack teeth on their oral jaws (de Graaf

2003). Most probably these species can only survive because specialised and more

efficient non-cyprinid piscivores are lacking. The piscivorous niche of these cooccurring

species is segregated by habitat, diet composition and prey size (de Graaf et

al., 2004). The main prey items eaten were B. humilis (40 % of the gut contents), B.

tanapelagius (32 %) and Garra species (21%). Therefore, the two small barbs form

the main link between the zooplankton and the piscivorous fish in the food web of the

Lake.

Besides piscivores, there are five other trophic groups of labeobarbs. One species

feeds on macrorophytes (L. surkis), one upon macrophytes and molluscs (L.

gorgorensis), one species on macrophytes and adults insects (L. osseensis), one

species predominantly on zooplankton (L. brevicephalus) and four species are

benthivorous feeding mainly on chironomid larvae and on macrofauna associated with

macrophytes (L. crassibarbus, L. intermedius, L. nedgia, L. tsanensis). If there is any

decline or reduction in the stock of large barbs, then the energy and the functioning of

the lake-floodplain-river ecosystem will be disrupted. It is obvious that unless strict

mitigation measures are taken the proposed dam will reduce the stock of labeobarbs in

Lake Tana and hence disrupt the food web therein.

Ribb River Fishery Studies: Final Report January 2008

Clarias Piscivorous Labeobarbus (7) Benthivorous

Labeobarbus

(3)

B.tanapelagius and B. humilis

Phytoplankton

Fig. 3. Simplified food web of Lake Tana

Gonad Maturity Status and Migration Behavior of Labeobarbus spp.

Gonad maturity stages were classified using keys as given in Table 2. From the total

of 2075 Labeobarbus specimens, only 297 were immature (gonad stages I, I1 and 111),

whereas 887 specimens were mature (gonad stages IV, V). Eight hundred thirty five

specimens of Labeobarbus species were running (gonad stage VI) and 53 were spent

(gonad stage VII). Immature gonads (I, 11,111) were relatively more numerous in case

of L. itltermedius as compared to other Labeobarbus species (Fig. 3). The data also

showed that about 85 % of the Labeobarbus species were either reproductively

mature or running. Fish with gonad stage V were only caught at the Ribb River mouth

in all the species of Labeobarbus (Table 7). As indicated in Table 7, the majority of

Labeobarbus specimens collected in the upstream sampling sites were running i.e.,

Ribb River Fishery Studies: Final Report January 2008

they shed their eggs and sperm when their abdomens were slightly pressed. No

running specimens were caught at the river mouth. All of the 53 spent Labeobarbus

fish were caught in the upper main channel and small tributary rivers of Ribb River.

Spent gonads were common in the samples collected at the end of October (Fig. 4).

Gonad development precedes spawning migration in most fish species and it is under

endocrine control from the pituitary gland and this gland requires a triggering

environmental factor (Payne, 1986). In the temperate zone, variation in day length is

a major triggering environmental factor for gonad maturation, but in the tropics, most

probably the variation is insufficient to be considered as a major factor (Wootton,

1990). However, like spawning period, the timing of gonad maturation must

generally coincide with the time of reasonable food supply, which means the fish

must lay down so much fat over the feeding phase for its survival and gonad

development (Payne, 1986; Wootton, 1990). Migration to the breeding area, for

tropical freshwater fishes, seems mainly triggered by rainfall patterns and water level

variations (Lowe-McConnell, 1975).

Ribb River Fishery Studies: Final Report January 2008

Fig. 4. Maturity stages in the most common Labeobarbus spp.

The absence of running (gonad stage VI) Labeobarbus fish at Ribb river mouth

(Table 7) shows that the river mouth is not a terminal spawning place, rather the fish

species aggregate there to start migration to upstream areas. More than 96% of the

specimens in the genus Labeobarbus in each particular species in the upstream area,

except L. intermedius, were either running or spent; the remaining were immature.

Ribb River Fishery Studies: Final Report January 2008

Table 7. Spatial distribution of gonad stages of the Labeobarbus spp. of Lake Tana

spawning in Ribb River.

Most large cyprinids of Africa spawn by making a single annual breeding migration

to upstream areas of rivers (Lowe-McConnell, 1975; Tomasson et al., 1984). This is

the best indication that they are not fully adapted to the lake environment. From the

previous studies conducted at four tributary river mouths (Gelgel Abbay, Gelda,

Gumara, and Ribb), this ancestral (riverine) reproductive strategy is found to be a

Samplin

g site

RM

MC I

CH

SH

KI

KE

MC2

BA

MC3

HA

MC4

ME

MC5

Total

Gonad

I

0

3

0

0

0

0

0

0

0

0

0

0

0

3

Maturity

I1

10

1

24

5

14

0

8

9

2

0

10

2

0

1

176

stages

I11

2

2

4

8

9

3

0

6

6

6

1

0

0

2

1

5

118

IV

0

1

0

2

2

4

7

0

5

0

0

0

21

v

8 6

7

0

0

0

0

0

0

0

0

0

0

0

867

VI

0

13

2

10

8

46

11

5

11

9

27

12

6

34

76

19

12

2 1

835

VII

0

1

5

4

0

2

1

7

5

1

2

0

3

0

4

53

Ribb River Fishery Studies: Final Report January 2008

characteristic for at least seven (L. acutirostris, L. brevicephalus, L. macrophthalmus,

L. megastoma, L. platydorsus, L. truttiforrnis, and L. tsanensis) of the 15 Labeobarbus

species of Lake Tana. The remaining 'missing' Labeobarbus spp. might possibly

migrate and spawn in other inflowing rivers (Dirma, Megech, and Arno-Garno), or

maybe even within the lake itself (lacustrine spawning) (Nagelkerke and Sibbing

1996; Palstra et al., 2004; de Graaf et al., 2005).

Gonad Maturity sl

Oct

Fig. 5. Temporal variation of gonad stages of Labeobarbus of Lake Tana in Ribb

River.

The migration pattern of Lake Tana's riverine spawners Labeobarbus species is

partitioned into three major phases:

(1) migrating from the foraging area of the lake to affluent river mouths;

(2) migrating upstream in the rivers' main channels; and

(3) entering a tributary for spawning after sunset.

Heavy rainfall usually starts in May and peaks in July and August in the Lake Tana

area (Tesfaye Wudneh, 1998; Eshete Dejen, 2003). During this time the tributary

rivers increase in volume and cause massive soil erosion. As a result of the inflow of

sediment and dissolved organic compounds, turbidity, increased water level, or a

combination of both is hypothesized to serve as environmental cues to trigger

Ribb River Fishery Studies: Final Report January 2008

spawning migration of Labeobarbus species to river mouths and upstream areas

(Sibbing et al., 1998).

Spawning Segregations

Spatial segregation

The relative contribution of each Labeobarbus within the sampling sites is shown in

Table 6. Two species of Labeobarbus (L. intermedius, and L. brevicephalus) were the

most abundant at the river mouth, in the main channel and tributary streams.

However, from all the upstream sampling sites of Ribb River, L. brevicephalus was

mostly common in the tributaries, particularly in Keha, whereas L. intermedius was

most abundant in the Ribb main channel, especially at the Bahir Dar- Gondar Bridge

site (Fig. 5). L. megastoma was common in the tributaries such as Barya, Keha and

Hamus Rivers but L. tsanensis and L, truttiformis were rare in the upper tributaries of

Ribb although they were commonly caught at the river mouth.

KE MC2 BA

Sampling sites

Fig. 6. Spatial segregation of Labeobarbus species of Lake Tana spawning in Ribb

River.

Ribb River Fishery Studies: Final Report January 2008 -

In spite of the fact that tropical cyprinids in general lack parental care and other

adaptations like viviparity or aestivation (Harikumar et al., 1994), they have certain

requirements in choosing their spawning places. Fast flowing, clear, highly

oxygenated water, and gravel-bed streams or rivers are preferred places for

Labeobarbus (Rodriguez-Ruiz and Granado-Lurencio, 1992). These conditions are

important for the growth of the larvae (Tomasson et al., 1984). Deposition of eggs in

the gravel or pebble beds protects the juvenile from being washed away by riffle, and

clear water will not prevent diffusion of oxygen. Previous studies (Alekseyev et al.,

1996; Nagelkerke and Sibbing, 1996; Dgebuadze et al., 1999; Palstra et al., 2004) on

spawning migrations of Lake Tana's Labeobarbus were focused on Gumara River as

it was considered as ideal breeding ground. On the other hand, Ribb River was not